Method and apparatus for manufacturing a component in layers

This application is the US national stage of International Patent Application No. PCT/EP2022/050482 filed on Jan. 12, 2022, which claims priority to German Patent Application No. 10 2021 101 984.2 filed on Jan. 28, 2021.

The present disclosure relates to a method and an apparatus for manufacturing a component in layers, in particular by dropwise application of a liquid material using a printhead.

Additive manufacturing processes are characterized by a high degree of design freedom and by tool-free manufacturing. They are therefore particularly suited for individual parts and components having a high degree of complexity that cannot be manufactured using conventional manufacturing processes, or can be manufactured only at great expense. In these additive manufacturing processes, the workpieces are built up in layers or in elements based on digital models.

There are a variety of additive manufacturing processes, using which a component can be manufactured in layers. Typical processes for building up metallic components are the process known as “Laser Powder Bed Fusion” (LPBF), the “Direct Energy Deposition” (DED) process and the so-called “Material Jetting” (MJT) process, in which molten material is printed directly onto a build platform by a printhead through one or more individually controlled nozzles.

An apparatus that dispenses individual drops based on a control signal is usually referred to as a “drop-on-demand” (DOD) printhead. An electrical signal is sent to the printhead actuator for each drop to be generated. Depending on the number of nozzles in a printhead, two different strategies, which are referred to as raster-based printing or vector-based printing, can be used. In raster-based printing using a printhead having more than one nozzle, the printhead is, in general, moved in a meandering pattern over the component. In vector-based printing having only one nozzle, the movement of the printhead is freely selectable. In both cases, the spacing of the droplets along the feed direction results from the speed of the printhead and the frequency of the trigger signal sent to the printhead actuator. Furthermore, in raster-based printing, the line spacing transverse to the feed direction of the printhead results from the arrangement of the nozzles in the printhead and the direction of movement. In the case of vector-based printing, this spacing results solely from the selected path.

In vector-based printing, the signal for the droplet generation can be generated in two ways. For example, a constant printing frequency can be used and then, in general, the path speed during droplet dispensing is also chosen to be constant, because otherwise a uniform trace cannot be generated. Furthermore, it is possible to measure or simulate the traversed path during the generation of the same and to generate a new trigger signal each time the desired droplet spacing within the path is reached. In this way, variable path speeds can be used; however, a correspondingly accurate and high-frequency measurement technique is required.

WO 2020/198038 A1 describes a method for monitoring a layer height in an additive manufacturing process. For example, optical or other means are used to measure layers in a layer-by-layer manner. To correct any errors that may occur, parameters such as a drop size, a drop spacing and other conditions can be modified.

It is one non-limiting object of the present disclosure to disclose techniques that make possible manufacturing of a component in layers, such that each layer may have a surface as flat as possible, a constant layer thickness and/or a pore volume as small as possible.

2 FIG. LV If one considers the geometry of a deposited droplet trace in additive manufacturing using the MJT process, its shape is highly dependent on various process parameters such as the droplet size, the speed of the droplets upon impact, the alloy composition, the temperature of the droplets and the substrate, one or more properties (material, structure, etc.) of the substrate, and the droplet spacing along the feed direction. As a result, neither the height nor the width of a trace is, in general, constant along the printing direction. As shown in, for example, there is in particular a rounding off of each trace at the end and at the beginning thereof, which can even be different. Furthermore, in the case in which multiple, possibly overlapping, traces are deposited along parallel paths, a layer produced in this way has a geometry which, in addition to the above-mentioned influences, also depends in particular on the spacing (D) between the paths.

3 FIG. 3 FIG. When building up multiple layers, the component edge is of particular interest. The surface tension of the liquid material used, for example a metal, enables the manufacture of overhangs; nevertheless, such an overhang has a negative effect on the flatness of the layer, because the component rounds off or sags toward the overhang. This is illustrated in an exemplary manner in. Even if no overhang is present, a similar problem can occur, as shown on the left in.

In one aspect of the present teachings, a hull-core strategy is utilized when building up three-dimensional components using the MJT process, in particular to reduce the rounding of edges and to produce layers that are as flat as possible. Thus, when printing the hull, compensation for overhanging component regions can be carried out. Furthermore, when printing the hull, idle-run loops can be used, in particular, to ensure a constantly high path speed during drop dispensing for paths having a large curvature or sharp corners.

In contrast to processes in which a measurement of the layer is required in order to make corrections if necessary, in the present invention the specific implementation is already known during path planning, and the strategies according to the invention require no additional equipment for the measurement and/or post-processing of the layers. Such additional equipment has significant disadvantages, especially in terms of manufacturing time, since it represents an additional process step. Moreover, it increases the degree of complexity of the machine, which has a detrimental effect on cost effectiveness. Another advantage of the present invention is that the compensation of overhanging regions of the component makes possible that more complex components having such overhanging regions can be manufactured without the use of support structures.

1 FIG. 1 FIG. 1 FIG. 10 14 10 100 10 100 102 16 104 102 10 106 102 104 102 104 106 102 104 10 104 10 102 In the MJT process, three-dimensional components are built up in layers in a known manner. For example,shows a portion of a componentconsisting of multiple layers. The componentis manufactured by an apparatusfor additive manufacturing of the component. For this purpose, as shown in, the apparatusincludes a printheadthat is configured to apply a liquid material, a moving apparatusthat is configured to move the printheadand the componentto be manufactured relative to each other, and a control devicethat is configured to control the printheadand the moving apparatus. In the example shown in, the build-up direction is the z-direction, and the printheadand the moving apparatusare moved relative to each other in the x-y plane in one layer and in the z-direction during the transition from one layer to the next. It is understood that the control devicecan be configured to either move the printheadwith respect to the moving apparatusand the componentdisposed thereon, or conversely to move the moving apparatushaving the componentdisposed thereon with respect to the printhead. A corresponding manufacturing technique is known to those skilled in the art, so that further details are not described herein.

100 10 12 The data for controlling the apparatusare generated in advance from a digital model of the component to be manufactured. This is done with the aid of suitable algorithms. Thus, in a first step, the component is divided into layers (e.g., i−3, i−2, . . . , i+2) perpendicular to the build-up direction. An outer perimeter of the componentis defined by a contour.

In a second step, the machine control data required to manufacture each individual layer are calculated. According to the invention, a hull-core strategy is used to build up an individual layer by the deposition of individual drops along calculated paths.

4 FIG. 20 14 12 10 18 20 12 12 12 shows an example of the path planning of a layer according to the invention when using the hull-core strategy. As used herein, the term “hull” means an outer regionof the layerthat extends along and/or defines the contourof the component. Furthermore, the term “core” means a central core regionof the layer that is at least partially surrounded by the outer region. In some embodiments, the hull completely surrounds the core, that is, it extends along the entire contour. However, it is understood that in other embodiments, the hull can be provided only along segments of the contour, for example, at the beginning and at the end of the respective core paths, and in particular need not be contiguous. It follows that in some embodiments, the core can define the contourat least in sections, for example in regions in which there are lower requirements for the flatness of the layer. Further, the hull need not have a constant width, which will be discussed in more detail below.

12 10 14 20 14 12 18 14 20 16 18 20 16 18 22 16 20 24 12 18 20 12 20 LV,core In general, in the process according to the invention, the contourof the componentto be manufactured is first defined in a layerthereof, based on the aforementioned data of the digital model of the component to be manufactured. Based thereon, the outer regionof the layerthat extends along the contour(the hull) and the central core regionof the layer, which is surrounded by the outer region(the core), are then determined. Then, the paths required to form the hull and the core are calculated. Thereafter, the dropwise application of the liquid materialin the core regionand in the outer regiontakes place along the calculated paths. In one example, the dropwise application of the liquid materialin the core regiontakes place along multiple parallel, for example rectilinear, core pathsthat have a first spacing Dfrom each other. Preferably, the paths are selected in accordance with the geometry of the component so that they have the greatest possible length. Furthermore, the dropwise application of the liquid materialin the outer regiontakes place along one or more outer pathsthat extend parallel to the contour. Thus, it is understood that the application of the material in the core region and in the outer region can take place in any order, that is, it may be started either with the core regionor with the outer region. Furthermore, the application along the respective paths need not necessarily be continuous. For example, sections of individual paths can be skipped in a first pass and can be filled in in one or more later passes. The core region need not be completely filled with material, i.e., recesses and the like can be provided in the core region. In this context, it is also apparent that the contouris not limited to an outer contour, but can also be an inner contour of the component. Furthermore, it is understood that two or more contours can also be present in a layer, so that two or more associated “outer” regionscan then also be present. The term “outer region” should therefore be understood broadly to indicate, in general, a region adjacent to a contour.

4 FIG. 20 22 LV,hull VS,hull HK VS,core As shown in, in the present example the hull (the outer region) consists of one to five, for example two, paths, which in the case of multiple paths each have a spacing Dfrom one another. In the feed direction (i.e. along the path), the droplets of the hull are deposited at a constant spacing D. The core has a spacing Dfrom the innermost path of the hull and is filled by the parallel core paths. Within the paths of the core, the droplets are deposited with a constant spacing D. Even though constant spacings in the feed direction are used in the present example, it is understood that varying spacings can be used in other embodiments, for example in accordance with different requirements in different regions of the hull and/or the core. The same applies in a corresponding manner to the spacings between the paths, for example, of the core.

4 FIG. 4 FIG. 12 12 102 10 14 102 102 102 BH HK As can be seen from, the respective paths of the hull and the core are calculated so that the diameter of the droplets is taken into account accordingly, firstly to obtain the desired contour, and secondly to obtain a constant layer height, in particular in the transition region between hull and core. This leads to the fact that, as shown in, the outermost path of the hull extends at the spacing Dfrom the contour, and likewise the innermost path of the hull extends at the spacing Dfrom the paths of the core. Furthermore, it can be seen that herein a line along which the printheadis moved with respect to the componentor the layerthereof is meant by the term “path”. The droplets are deposited in a corresponding manner along the path, wherein the start point and the end point of a path correspond to the center point of the first and the last deposited droplet, respectively. Depending on the position of the printheadalong a respective path, the printheadis controlled in a suitable manner to deposit the individual drops with the desired spacing. Thus, it is understood that this is done in accordance with the path speed of the printheadalong the respective paths.

18 20 20 18 hull core The deposited droplets have a (an average) diameter d, which is determined, in essence, by the amount of material dispensed per droplet and the physical properties of the material. Thus, in some embodiments, different average droplet diameters are used for the core regionand for the outer region, respectively, by suitably varying, for example, the amount of material dispensed per droplet. In particular, in the outer region, the average droplet diameter dcan be less than or equal to the average droplet diameter din the core region. Droplet diameters of individual paths in the core region and/or in the outer region can have different sizes relative to each other.

2 FIG. 2 FIG. 30 32 22 22 30 32 22 12 10 12 Layer HK BH The advantage achieved by the hull-core process will be explained once again with reference to. For example,can be viewed as illustrating a portion of an exemplary core region within a layer. As was mentioned above, it can be seen that, for example, at the beginningand at the endof each core path, a rounding occurs that varies depending on the direction in which the core pathis traversed. This is a result of the dropwise application of the liquid material as the printhead moves. This leads to the different height (H) of the layer shown in the right part, in particular also at the beginningand at the endof the respective core paths. By providing the hull along the contourof the componentaccording to the invention, these differences can be compensated. As will be explained again below, the spacing between the hull and the core D, the spacing Dbetween the outermost path of the hull and the contour, etc., for example, play a role here.

24 18 22 24 24 HK LV,core In some embodiments, the innermost one of the outer pathsis calculated so that it has a second spacing Dfrom the core regionthat is smaller than the first spacing Dbetween the core pathsfor at least a portion of the innermost path. In this way, a suitable overlap between the outermost droplets of the core and the droplets of the innermost pathof the hull can be created or enlarged.

BH LV,core 24 12 22 12 In addition, it can be advantageous to define a third spacing Dbetween an outermost one of the outer pathsand the contoursuch that it is smaller than the first spacing Dbetween the core paths, for example about half as large, to ensure that the contouris maintained in this way.

16 12 HK BH hull Therefore, it is understood that the droplets of liquid materialhave the average diameter d, and the second spacing Dand/or the third spacing Dcan be approximately 0.4 to 0.6 times the average diameter d (for the case that differing diameters for the hull and core are used, the average droplet diameter din the hull). In this way, as already explained, a layer as flat as possible can be obtained, and the contourcan be well maintained. As mentioned above, in some embodiments, droplets having different average diameters can be used in the core and in the hull. Droplets having a diameter (of the deposited droplets) between, for example, 200 and 1000 μm can be used.

VS,hull VS,core VS,core VS,hull LV,hull LV,core 24 22 24 In addition, it can be provided that a fourth spacing Dbetween adjacent droplets along the path on one of the outer pathsis about 0.6 to 1.1 times the average diameter d (of the hull), and/or a fifth spacing Dbetween adjacent droplets on one of the core pathsis about 0.5 to 1 times the average diameter d (of the core). In particular, it is provided that the fifth spacing Dis different from the fourth spacing D; for example, it is 0.7 to 0.95 times the same. Similarly, a sixth spacing Dbetween adjacent ones of the outer pathscan be determined so that it is 0.8 to 1.2 times the average diameter d of the droplets (in particular, of the hull). In addition, the first spacing Dcan also be 0.8 to 1.2 times the average diameter d of the droplets (of the core). In this way, the densest and flattest layer can be obtained.

24 22 It has also been found to be advantageous, for example, if at least two adjacent outer pathsare traversed in opposite directions, and/or at least two adjacent core pathsare traversed in opposite directions. In this way, systematic deviations that can occur when one direction is traversed can be compensated for.

2 FIG. 4 FIG. 4 FIG. 30 32 18 24 30 22 32 22 14 HK HK As explained with reference to, the rounding at the beginningand at the endof each path can be different. Therefore, it can be advantageous if the second spacing Dbetween the core regionand the innermost of the outer pathsis varied so that it is different at the beginningof each core paththan at the end. For example, if the core pathsinare all traversed in the same direction, the spacing Din the section shown incould be selected so that it is greater or less than the spacing on the opposite (not shown) side of the layer.

22 24 14 10 In some examples, it can also be advantageous if the core pathsand/or the outer pathsare traversed in two successive layersin different directions. In this way, the height of the componentcan be obtained as uniformly as possible. Additionally or alternatively, the paths of the hull in particular can start in successive layers at different, for example randomly selected, points.

By suitably choosing the above-explained spacings between the individual drops and paths, layers that are as flat as possible can be obtained according to the geometry of the component, without the need for corrections to compensate for an uneven layer height following completion of a layer.

3 FIG. 3 FIG. 11 10 14 12 14 As was already mentioned in the introduction, it is also possible to produce overhanging contours (also referred to as “downskin”) using the LMP process. This is shown in an exemplary manner in. As can be seen in, such an overhangof the componentis described, for example, by a first angle α to the build-up direction within a layer. It is understood that the first angle α can vary in the build-up direction, so that it can, in essence, be specified by the angle between the tangent to the contourin a given layerand the build-up direction.

3 FIG. 5 FIG. 5 FIG. 5 FIG. BH 24 12 14 11 24 24 18 12 11 24 12 As also shown in, the overhang can have a negative effect on the flatness of the layer, since the component can round off towards the overhang and sag. However, in accordance with the invention, it has been found that this can be counteracted by increasing the third spacing Dbetween the outermost of the outer pathsof the hull and the contourin the layerin the region of the overhangby a compensation spacing Δ, based on the first angle α. This is shown as an example in. In, dotted lines show the path′ without correction that would be used if there were no overhang, while the pathactually used is corrected so that it extends further inward, that is, toward the core region, by the compensation spacing Δ. It can be seen that, in the example shown in, the contourhas the overhangonly in the area indicated by the solid line, while no correction of the spacing of the outermost of the outer pathsis made in the remaining area of the contour.

24 24 24 18 HK It has proved to be advantageous to select the compensation spacing Δ so that it is zero for an angle of 0° to the build-up direction and is increased as a function of the first angle α. Thus, for example, the spacing can be increased by 2 to 20 μm per degree. In other words, the greater the overhang, the further inward the outermost of the outer pathsis moved. It will be understood that, in some embodiments, multiple, or even all, of the outermost pathsof the hull may be moved inwardly. Thus, it can also be necessary or useful to reduce the second spacing Dbetween the innermost of the outermost pathsand the core region, for example, by the compensation spacing Δ.

6 FIG. A further advantageous embodiment of the method described herein is described below with reference to.

6 FIG. 12 14 13 41 42 102 41 42 13 10 As shown in, the contourin the layercan have a regionin which the contour has a local curvature that is greater than a predetermined value. In particular, the local curvature is thereby denoted by the curvature of the path between two successive drops,in the region that, for example, the printheadmust traverse in order to deposit the successive drops,. In particular, the regionmay correspond to a corner of the componentto be manufactured. At such a corner, for example, an angle R can be present between adjacent legs of a polygon shape. In such a case, the angle R can also be used as a measure for the curvature.

102 104 Since paths cannot be traversed at sufficiently high path speeds in regions of high local curvature due to the limited axis dynamics of the printheador the moving apparatus, lesser path speeds must be used in some circumstances, resulting in lower drop generation frequencies. However, this results in reduced energy input and thus local cooling of the component, which leads to undesirable deviations in the generated geometry, in particular at corners and the like. It has therefore proved to be advantageous to stop the drop generation at high path speed in such regions, to traverse a loop without applying drops, and to reach the corner point again at high path speed in the desired direction. Then droplet generation can start again. In this way, all drops can be applied at a substantially constant high speed.

6 FIG. 13 41 27 27 12 16 27 12 42 102 104 27 41 24 42 24 12 13 VS,hull VS,hull In the example shown in, for example, the application of the individual drops takes place from left to right. This means that, upon reaching the region, a final dropletis applied at high path speed, and then an auxiliary pathis traversed, wherein the application of the liquid material is stopped. In particular, the auxiliary pathis a closed auxiliary path outside the contour, which is traversed without applying liquid material. The auxiliary pathis calculated in a suitable manner so that the portion of the contourat which the next dropis to be applied can be reached at the desired speed in as short a time as possible. This is done using suitable optimization algorithms (for example, by nonlinear optimization using a known interior-point method), wherein, for example, a maximum relative speed between the printheadand the moving apparatus, a maximum relative acceleration, or a maximum temporal change in the relative acceleration are used as constraints in an optimization with respect to the time required to traverse the auxiliary path. In particular, jerking of the printhead or the moving apparatus due to large accelerations and/or changes in acceleration is to be avoided. Over the course of this, a position of the last dropleton the outer pathbefore stopping of the application and, based thereon, a position of the first dropleton the outer path after continuation of the application can further be determined so that a direct spacing f*Dbetween the two droplets is a predetermined spacing. For example, this spacing can be 0.8 to 1.2 times the spacing Dbetween adjacent drops in sections of the outer pathin which the contourdoes not include the region.

14 Using the methods described above, the present invention makes it possible to obtain a layer thickness that is as uniform as possible within each layer, in particular even when an overhang is formed, without having to perform additional measurements or provide additional devices. This is achieved by calculating in advance all of the paths to be traversed in one layer, taking into account the partition into hull and core, depending, if necessary, on whether there are overhangs or corners of the component.

10 In a simplest case, in which the componenthas no overhang and no regions of high local curvature, and the core area is completely filled with material, the outer region can be defined, for example, so that it completely surrounds the core region and has a desired width. Then, in a first step, the core region can be traversed along parallel paths until it is completely filled. Following this, the outer region is then traversed, for example along one to five outer paths, which have predetermined spacings from one another or from the contour and from the core region. Following this, it can proceed with the production of the subsequent layer.

It is explicitly emphasized that all features disclosed in the description and/or claims are to be considered separate and independent from each other for the purpose of the original disclosure as well as for the purpose of limiting the claimed invention regardless of the combinations of features in the embodiments and/or claims. It is explicitly stated that all range indications or indications of groups of units disclose all possible intermediate values or subgroups of units for the purpose of the original disclosure as well as for the purpose of limiting the claimed invention, in particular also as a limit of a range indication.